RU2616764C2 - Device and method for evaluation of heart rate while moving - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: inventions refer to medicine. The method for determination of human heart rate is implemented by means of a portable device, which is part of the system to determine the heart rate. The portable device for human heart rate determination contains a heart rate measuring unit for generating a of heart rate signal, a movement measuring unit for measurement of human body parts movement to generate a motion signal, and a processing unit for heart rate signal frequency quality, heart rate calculation based on the heart rate signal if the signal quality is above the predetermined threshold, and evaluation of heart rate based on the motion signal, if the signal quality is below the said threshold. The processing unit evaluates the heart rate signal based on the motion signal by estimating a constant heart rate HR_constant and determining the exponential changes in heart rate over time. The exponential change in heart rate starts from the last valid measured heart rate and ends at an estimated frequency of HR_constant, which depends on the motion signal frequency. The last reliably measured heart rate value is the last heart rate measured by the heart rate measuring unit at the time before reaching the said threshold.

EFFECT: invention provides increased accuracy of heart rate determination.

13 cl, 12 dwg

Description

FIELD OF THE INVENTION

The present invention relates to a portable device for determining a person’s heart rate. The present invention also relates to a corresponding method and system. In addition, the present invention relates to a computer program for controlling said device so as to perform the steps of said method.

BACKGROUND

Due to the growing number of people who live inactive lives, over the past decades many products and services have been developed to promote physical activity, both for research and commercial purposes. The mentioned products of promoting physical activity in most cases try to calculate or estimate the heart rate in order to display the person’s pulse during physical activity. Currently, the most successful devices that measure heart rate for athletes use a chest belt. These devices measure the electrical signal of the heart (ECG) during an athlete's physical activity. However, these chest straps are uncomfortable to wear, which actually limits their use by serious athletes.

As more people are aware of the importance of controlling heart rate for their health, and most people are trying to avoid wearing this kind of chest strap because of their inconvenience, the paradigm of measuring heart rate is slowly changing from high resolution and low comfort to medium resolution, but with higher wearing comfort.

This is achieved, for example, by means of optical heart rate monitors, which can be attached to various parts of the body, for example, also to the athlete’s wrist. A device of this kind, which is known in the art, is commercially available under the name ePulse2 ™. This heart rate monitor includes an optical sensor that is similar to the pulse oximeters available on the market. It is implemented as a wristband that can be comfortably worn on the wrist.

The optical sensors used to measure heart rate, however, suffer from large movement artifacts, especially in the case of large and fast movements that occur during physical activity such as running, cycling or rowing. This is because the optical sensor actually optically measures blood flow in the blood vessel, and of course, body movements also influence the blood flow, so intermittent, gross movements occur within the blood vessel. This leads to large movement artifacts that complicate the measurement of heart rate.

For this reason, some optical sensors known in the art use an additional motion sensor in order to measure the ongoing movement of a part of the body and compensate for the resulting artifacts of movement. However, there is a limit. When the movement of the body part to which the sensor is attached becomes very large, the optical sensor no longer provides reliable measurements, even when the measured signal is compensated and accordingly adapted using the motion signal provided by the motion sensor.

In this case, the heart rate monitor either displays an incorrect heart rate value or does not show any heart rate value at all. This is considered a major drawback, as the measurement gives an inaccurate heart rate value or does not give any heart rate value at all.

Years of experience with a large group of users have shown that participants attach great importance to the reliability and comfort of such heart rate measurements. In particular, when athletes or athletes are subjected to severe stress or perform heavy exercises, it is necessary to have reliable operational information about the current heart rate at any given time. If a heart rate monitor has detected an abnormal heart rate or has not even shown a heart rate value, this can be perceived as demotivation and has a negative effect on the overall perception of the device.

SUMMARY OF THE INVENTION

An object of the present invention is to provide an apparatus, method, system and associated software of the kind that was originally mentioned that provides an improved measurement of heart rate, in which the heart rate is measured in a convenient manner for the user, and the measurement gives reliable results even when sudden movements occur in the measured part of the user's body. In particular, it is an object of the present invention to calculate or estimate the heart rate with the highest degree of accuracy and overcome the problem of large artifacts in the heart rate signal, which may make it impossible to measure the heart rate.

In the first aspect, this task in accordance with the present invention is solved by a portable device for determining the heart rate of a person, containing:

a heart rate measuring unit for measuring a person’s heart rate over time in order to generate a heart rate signal,

- a motion measuring unit for measuring the movements of a part of the human body over time in order to generate a motion signal, and

- a processing unit that is configured to measure the quality of the heart rate signal, calculate the heart rate based on the heart rate signal if said signal quality is above a predetermined threshold, and estimate the heart rate based on a motion signal if said signal quality is below said the threshold.

In a second aspect of the present invention, there is provided an appropriate method, which includes the steps of:

- measuring a person’s heart rate over time in order to generate a heart rate signal,

- measuring the movements of a part of the human body over time in order to generate a motion signal,

- measuring the quality of the heart rate signal, and

calculating a heart rate based on a heart rate signal if said signal quality is above a predetermined threshold, and estimating a heart rate based on a motion signal if said signal quality is below said threshold.

In a third aspect of the present invention, there is provided a system for determining a human heart rate, comprising:

a portable heart rate measuring device for measuring a person’s heart rate over time to generate a heart rate signal,

- a portable motion measuring device for measuring the movements of a part of the human body over time to generate a motion signal, and

- a processing device that includes a communication interface for receiving said heart rate signal and said motion signal, and processing means that is configured to measure the quality of the heart rate signal to calculate a heart rate based on the heart rate signal, if said signal quality is above a predetermined threshold, and estimates of heart rate based on a motion signal, if the quality mentioned during signal below said threshold.

In an additional aspect of the present invention, there is provided a computer program product comprising software code means causing a computer to control said portable device so as to perform the steps of said method when said computer program is executed on a computer.

Preferred embodiments of the present invention are defined in the dependent claims. It should be understood that the claimed method and the claimed system have similar and / or identical preferred embodiments as the claimed portable device, as defined in the dependent claims.

The inventors have found that the heart rate can be estimated reliably even when the heart rate signal generated by the heart rate measuring unit becomes unreliable due to motion artifacts. In accordance with the present invention, the heart rate in such cases is estimated based on a motion signal provided by the motion measuring unit. An estimate of the heart rate based on the motion signal means that the heart rate is estimated based on at least the motion signal, which in turn means that other parameters and signals can also be included in this estimate. In order to implement this type of heart rate measurement, the processing unit is configured to measure the quality of the heart rate signal. When the heart rate signal is above a predetermined threshold, the heart rate can be measured based on the heart rate signal. This can be done by estimating the frequency of the measured heart rate signal, which leads to the determination of the human heart rate.

On the other hand, if it is found that the signal quality of the heart rate is lower than the predetermined threshold, the heart rate can be estimated based on the motion signal. For this purpose, the processing unit is configured to switch from a first mode in which a heart rate is calculated from a measured heart rate signal to a second mode in which a heart rate is estimated based on a measured motion signal. Switching between these two modes depends on the threshold value.

Said threshold value indicates the noise level in the heart rate signal at which the heart rate signal measured by the heart rate measuring unit becomes unreliable when a certain noise level is exceeded, and accordingly when the signal reaches the threshold. In other words, the threshold value indicates the minimum quality of the heart rate signal, which is necessary in order to reliably calculate the heart rate based on the heart rate signal. The threshold value can be determined from experiments that evaluate at what level the device’s movement artifacts appearing in the heart rate signal become so large or strong that a reliable heart rate measurement can no longer be extracted from the heart rate signal. This threshold or noise threshold value need not be an accurate value. It can also be a range of values where the quality of the heart rate signal moves from a good or sufficient measurement quality to a low, insufficient measurement quality. In other words, the threshold value indicates the lowest level of quality of the heart rate signal, below which the calculation of the heart rate based on the heart rate signal would lead to failure, and accordingly, an incorrect heart rate value that is outside the allowable error range.

Using a motion signal, one can reliably estimate the heart rate also in those cases in which the heart rate signal is unsuitable due to artifacts of large movements. This is a major advantage compared to the prior art devices mentioned above, as the user receives reliable information about his heart rate / heart rate even when the device is subjected to high accelerations or strong vibration.

Unlike prior art devices, the presented portable device thus allows to continuously provide the user with a heart rate, even at those times when the heart rate measuring unit stops working.

Estimating a heart rate based on a motion signal does not necessarily mean that only a motion signal is taken into account in order to estimate the heart rate. Even in the above cases in which the quality of the heart rate signal is below a predetermined threshold, the heart rate signal can still be taken into account. In such an embodiment, the processing unit is configured to correct and accordingly adapt the heart rate signal based on information taken from the motion signal. In other words, the heart rate signal in this case is corrected with correction values that can be determined from the motion signal, for example, filtering the noise that occurs in the heart rate signal by comparing / subtracting parts of the signal caused by the movement from the heart rate signal .

Even though the heart rate can be directly calculated from the heart rate signal, if the quality of the heart rate signal is above the predetermined threshold, the processing unit in accordance with one embodiment of the present invention can be configured to calculate the heart rate based on heart rate signal and adapt the calculated heart rate based on the motion signal. However, this is not necessarily necessary when the heart rate can be measured reliably. However, this measure can be implemented as an additional improvement in measuring heart rate.

In accordance with the present invention, a heart rate measuring unit may be implemented by any kind of sensor that allows a person to continuously measure a person’s heart rate over time. The sensor may also include an ECG electrode. According to a preferred embodiment of the present invention, the heart rate measuring unit includes an optical sensor, in particular a photoplethysmographic (PPG) sensor, for continuously measuring a pulse wave of human blood over time so as to generate a heart rate signal. The PPG sensor includes a photo detector, which usually measures the spectral absorption of blood at different wavelengths, allowing you to determine the absorption of light due to pulsating arterial blood.

This type of PPG sensor allows you to measure a person’s heart rate in a convenient way. The portable device may, for example, be attached to the wrist of a person. Unlike the well-known PPG sensors, which are usually attached to the tip of the finger or to the earlobe of a person, attachment to the wrist of a person allows the use of a portable device for various types of sports activities in which the device can be comfortably worn. Even though attachment to the wrist is preferred, the portable device can also be attached to any other part of the human body, for example, to the chest, leg or around the neck.

The aforementioned motion measuring unit preferably comprises an inertial sensor for measuring the acceleration of the body part to which it is attached in at least one spatial dimension. This inertial sensor is preferably configured to perform triaxial accelerometry. For this purpose, it is preferably provided with three accelerometers and / or three gyroscopes. Accelerometers are placed so that their measuring axes are perpendicular to each other so that inertial forces can be measured in all three spatial dimensions. Three gyroscopes are placed in a similar perpendicular way, which allows you to measure the rotational position of the device relative to an arbitrarily selected coordinate system. It should be understood that gyroscopes are not necessarily necessary, since accelerometers measuring acceleration in various spatial directions are sufficient for most devices. In addition, it should be noted that a single accelerometer is also sufficient to generate the desired motion / acceleration signal.

The portable device is preferably designed in a form that is similar to a watch. In accordance with one embodiment of the present invention, the portable device includes a display for displaying the calculated heart rate. This display provides the user with a measured heart rate / heart rate in real time. This display can be implemented in various ways, for example, as an LED matrix.

In accordance with one embodiment of the present invention, the processing unit is configured to determine the quality of the heart rate signal in the frequency domain by analyzing the spectral peaks of the heart rate signal at the heart rate and / or at its harmonics.

This analysis examines the height, and accordingly the magnitude of these peaks. This gives a definition of the strength of the heart rate signal. In general, it can be argued that the higher and sharper the peaks in the frequency domain, the better the quality of the heart rate signal. This is mainly based on the fact that the pulse wave of blood ideally produces a periodic signal, which in the frequency domain gives clear peaks at or near the heart rate and / or at the frequencies of its harmonics.

A clear peak at the heart rate and / or at the frequencies of its harmonics is thus an indicator of the periodic signal, which in turn is an indicator of good quality of the heart rate signal. If, on the other hand, the heart rate signal is damaged by the ongoing movement and includes motion artifacts, then this will lead to various noise peaks in the power spectrum. Generally speaking, signal quality in this way can be determined based on the spectral peaks of the heart rate signal.

If sharp peaks are observed at or around the heart rate and / or at its harmonics, the signal quality is reliable enough to calculate the heart rate based on the heart rate signal (first mode). If, on the other hand, the spectral analysis of the heart rate signal shows a noisy spectrum, the processing unit switches to a second mode in which the heart rate is estimated based on the motion signal.

In one embodiment, the portable device further includes a frequency filter for filtering out those frequency components in the heart rate signal that are generated due to movement of the device in which the processing unit is configured to determine the quality of the filtered heart rate signal. This filter provides easier detection of heart rate in a heart rate signal. However, this function is not necessary, since in practice the frequency components that are formed due to movement are formed at frequencies different from those frequency components that are formed due to heartbeats. Thus, in practice, in most cases, it is possible to clearly distinguish between different types of frequency components, especially when analyzing a heart rate signal in the frequency domain.

Instead of analyzing the heart rate signal in the frequency domain, the processing unit can also be configured to determine the quality of the heart rate signal in the time domain by analyzing the peak heights as a function of autocorrelation with the heart rate period and its multiple periods. In the case of analyzing a heart rate signal in the time domain, the processing unit measures the signal quality level, depending on the periodic components in the signal, which in turn are an indicator of the reliability of the heart rate signal. This can be done, for example, by counting the number of zero-line intersections and / or signal peaks, or by analyzing the density of signal peaks. In the same way as mentioned above, the processing unit then determines, depending on the signal analysis, whether to calculate the heart rate based on the heart rate signal (first mode) or to estimate the heart rate based on the motion signal (second mode).

An estimate of the heart rate based on the motion signal in the second operating mode of the processing unit is preferably performed as follows:

In accordance with one embodiment of the present invention, the processing unit is configured to estimate a heart rate based on a motion signal by estimating a heart rate _constant (HR _constant ) and determining an exponential change in heart rate over time, wherein an exponential change in heart rate begins with the last reliably measured heart rate and ends at an estimated value of HR _constant . The HR _constant value is the estimated human heart rate, which depends on the frequency of the motion signal, and the last reliably measured heart rate is the last heart rate measured by the heart rate measuring unit at the point in time before the mentioned signal quality level reaches threshold value.

The processing unit thus estimates the heart rate in two stages. At the first stage, the value of HR _constant is estimated. Assessing the value of HR _constant requires an assessment of the existing movement and, accordingly, an assessment of the frequency of movement of the device (the frequency of movement of the measured part of the body). This frequency can be obtained from the measured motion signal.

The HR _constant value indicates the level of the person’s heart rate to which the person’s heart rate would increase or decrease if the amount and intensity of movement would remain constant over time. In other words, the assessment is based on the assumption that the person’s movement is kept constant during the so-called transition period, in which the device switches to the second mode, in which the heart rate is estimated based on the motion signal. Since the transition period in practice is very short, not more than a few seconds, this assumption shows a good approximation as a result.

Once the HR _constant value is determined, the processing unit determines an exponential change in heart rate over time with an initial value equal to the last reliably measured heart rate that was measured based on the heart rate signal and with a final value equal to HR _constant . It has been shown that exponential change reflects the natural behavior of the human heart in a good and accurate way.

This exponential behavior is an approximation curve for increasing / decreasing a person’s pulse. This increase / decrease also depends on the physical fitness of the person. Thus, the exponential approximating curve preferably has a time constant a, wherein a is a constant related to the physical fitness of a person, which decreases with increasing physical fitness.

In accordance with one embodiment of the present invention, the processing unit is configured to calculate the HR _constant value according to the formula HR _constant = 2.1 * fa, where f is the frequency of the movement signal, and a is a constant indicating a person’s physical fitness according to the formula a = 75-HR _rest , where HR _rest is the resting heart rate of a person.

It should be noted that the above relationship between HR _constant and the intensity of a person’s physical activity, which is expressed by the frequency of movement of the measured part of the body, is a linear relationship. This ratio was found based on experiments that were performed by the applicant. These experiments showed that the value of HR _constant remains at a level that is approximately two times the frequency of movement. As a good approximation for the parameter of physical fitness a, the ratio a = 75-HR _rest was found.

The HR _rest resting HR value can, for example, be measured directly using the presented portable heart rate measuring device when the user is resting, that is, when the user is not moving. In addition, the HR _rest value can also be estimated from the heart rate signal during regular heart rate measurement (in the first operating mode of the processing unit). However, it should be noted that other values for the parameters a, HR _rest and HR _constant can also be selected without leaving the field covered by the present invention.

Instead of measuring or evaluating HR _rest , it is also possible for the portable device to include an input interface that can be implemented, for example, as a small keyboard or touchpad that allows the user to manually enter his / her HR _rest value. Similarly, it is also possible for the user to directly determine his / her personal fitness parameter a.

Instead of manually determining the fitness parameter a and measuring the frequency f, a much better way to determine the value of HR _constant is to use previous measurement sessions for the same user.

In accordance with one embodiment of the present invention, the portable device further includes a storage unit that is configured to store control measurements of heart rate for known levels of intensity of a person’s physical activity, the processing unit being configured to determine a level of intensity of a person’s physical activity based on the generated motion signal, and with the ability to determine a constant heart rate ( HR _constant ) by comparing a certain level of intensity with the control measurements stored in the said storage unit, if the signal quality of the heart rate is lower than the threshold signal quality.

Said control measurements may be heart rate measurements that were recorded during previous measurement sessions of the same user. If the user performed physical activity, such as, for example, running, on the eve of using the same portable device, then the measured signals can be stored in a storage unit. By recording the heart rate signal together with the corresponding motion signal, the calculated heart rate can be displayed / linked to the corresponding levels of motion that are obtained from the motion signal.

Thus, using the previous measurements, it is also possible to determine various constants of the heart rate, which correspond to different levels of intensity of a person’s physical activity. For example, it is possible that a certain number of intensity levels are displayed in the storage unit on the corresponding values of the heart rate constant. In this case, it is not necessary that all previous measurement data is stored in a storage unit.

The intensity level of a person’s physical activity can, for example, be determined based on the frequency of at least one peak value of the generated motion signal and / or on the basis of the average level or amplitude of the generated motion signal over a time interval. The intensity level thus indicates the level of movement, which is a measure of the physical exertion that a person is exposed to during his / her physical activity.

The storage unit may be implemented, for example, with a small microchip. Recording heart rate control or heart rate control constants can be done automatically. For this purpose, the processing unit switches to recording mode, in which the processed heart rate and the corresponding levels of movement intensity are simultaneously stored in the storage unit during measurement. The determination of the heart rate constant based on the stored control measurements, as explained above, leads to an increase in the measurement efficiency. The user no longer has to manually enter the parameters that are used to calculate HR _constant (parameters a, f).

The personalized HR _constant value can be determined efficiently by analyzing the previous recorded data. The obtained parameters are thus personalized, so their use improves future estimates of heart rate during transition periods when the processing unit switches to the second operating mode.

In accordance with a further embodiment of the present invention, the portable device further comprises a first input interface for receiving information about a type of human physical activity, the processing unit being configured to estimate a heart rate based on a movement signal and a type of physical activity if the quality of the heart rate signal is lower the mentioned signal quality level.

This first input interface can, for example, be implemented as a small keyboard that integrates into a portable device. Thus, the user can manually select the type of physical activity that he wants to perform. The user can, for example, show a selection list of various physical activities, such as, for example, running, cycling, rowing, weightlifting, etc.

If the type of physical activity is known in advance, this simplifies the determination of the frequency of movement, since each type of physical activity in practice generates its own characteristic type of structure of the detected motion signal with a different expected average frequency of movement. The presence of at least crude information about the expected frequency of movement allows you to save processing time in the second mode of the processing unit, in which the heart rate is estimated based on the motion signal.

Information on the type of physical activity may also allow adaptation of the aforementioned exponential approximating curve within the transition period. This is consistent with the fact that the heart rate can vary in time for different types of physical activity, for example, the heart rate can increase / decrease faster when riding a bicycle compared to weight lifting.

In accordance with a further embodiment of the present invention, the portable device comprises a second input interface in order to obtain personal data of a person, in particular age, gender, weight, height and / or heart rate at rest, the processing unit being configured to estimate the frequency heart rate based on the motion signal and personal data obtained if the quality of the heart rate signal is lower than the aforementioned threshold signal quality.

Instead of determining personalized parameters from previous measurements, as explained above, it is also possible for the user to manually enter his / her personal data, which physiologically affect the heart rate and its change over time. Said second input interface may be either an additional keyboard, or may be implemented by the same keyboard that is used as the first input interface.

Personal physiological user data can be used either to adapt the aforementioned parameters to calculate HR _constant , or to apply an additional physiological model, which can then be used to adapt the aforementioned model of changes in heart rate during the transition period. Examples of such physiological models are known in the art. Some exemplary models, for example, are known from the scientific publication “Reliability and Validity of the Combined Heart Rate and Movement Sensor Actiheart”, European Journal of Clinical Nutrition (2005) 59, 561-570.

In fact, the presented portable device and the corresponding method make it possible to detect the heart rate of an athlete or athlete in many different situations, even in situations where large movements occur, and optical heart rate sensors known in the prior art are not able to reliably determine the heart rate abbreviations. As explained above, the processing unit is configured to switch between two modes. In the first mode, the heart rate is calculated based on the heart rate signal if the signal quality is above a predetermined threshold. On the other hand, if the processing unit is not able to reliably calculate the heart rate based on the heart rate signal, the processing unit switches to a second mode in which the heart rate is estimated from movement by using a model in which data is compared with a set of values, determined from previous measurements or entered by the user. The presented method, thus, uses one of the following features or any combination of them:

1) a history of good measurements right down to the point of failure (unreliable measurement);

2) an indication of the quality of the measurements taken, which indicates that the last measurement is unreliable;

3) a user profile, which can be either pre-recorded or determined from reliable measurements; and / or

4) a physiological model that uses parameters determined from previous measurements for the same user.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other aspects of the present invention will become apparent from and will be explained with reference to the embodiments described below.

Figure 1 shows schematically the use of a portable device in accordance with the present invention.

Figure 2 shows a block diagram illustrating components of a portable device in accordance with the first embodiment.

Figure 3 shows a schematic diagram illustrating the input and output signals of a processing unit of a portable device in accordance with the first embodiment.

FIG. 4 shows a block diagram illustrating components of a portable device in accordance with a second embodiment.

FIG. 5 shows a first example of a measured heart rate signal (FIG. 5A), including a corresponding measurement of signal quality (FIG. 5B).

6 shows a second example of a measured heart rate signal (FIG. 6A), including a corresponding measurement of signal quality (FIG. 6B).

Fig. 7 shows a third example of a measured heart rate signal (Fig. 7A), including a corresponding measurement of signal quality (Fig. 7B).

FIG. 8 shows a first example shown in FIG. 5, including a heart rate signal that has been evaluated by the present portable device in accordance with the present method.

FIG. 9 shows a second example shown in FIG. 6, including a heart rate signal that has been evaluated by the present device in accordance with the present method.

FIG. 10 shows a third example shown in FIG. 7, including a heart rate signal that has been evaluated by the present device in accordance with the present method.

11 schematically shows the application of the system in accordance with the present invention.

12 shows an example diagram illustrating the relationship between heart rate and exercise power / intensity.

DETAILED DESCRIPTION OF THE INVENTION

Figure 1 schematically shows the use of a portable device in accordance with the present invention, which is indicated by a reference numeral 10. A person 20, which is shown here as an example by a runner, carries a portable device 10 for measuring heart rate during physical activity.

The portable device 10 is attached to a body part 12 that is suitable for measuring the pulse of a person 20, that is, to that part of the body 12 on which an arterial blood pulse can be easily tracked. As shown in FIG. 1, the portable device is preferably attached to the wrist of a person 20. However, the portable device can also be attached to any other part 12 of the human body 20, for example, to the chest, leg, or around the neck.

As shown in the schematic block diagram of FIG. 2, said portable device includes a heart rate measuring unit 14, a motion measuring unit 16 and a processing unit 18. The heart rate measuring unit 14 and the motion measuring unit 16 are electronically coupled to the processing unit 18. The heart rate measuring unit 14 preferably includes an optical sensor, in particular a photoplethysmographic (PPG) sensor, which continuously measures a pulse wave of human blood Single 20 for a time and generates a signal 22 of heart rate.

The PPG sensor includes a photodetector (not shown) that measures the spectral absorption of blood at various wavelengths, which allows you to determine the changes in light absorption that occur due to pulsating arterial blood. This type of PPG sensor allows you to measure a person’s heart rate in a convenient way.

The motion measuring unit 16 preferably includes an inertial sensor to measure the acceleration of said body part 12 in at least one spatial dimension, more preferably in all three spatial dimensions. This inertial sensor continuously measures the movement of said body part 12 of the person 20 in time to generate an acceleration signal in time that records the accelerations occurring on the wrist of the person 20, to which the portable device 10 is preferably attached.

Thus, the processing unit 18 receives two signals, which are measured in real time, a heart rate signal 22 and a motion signal 24. This is approximately shown in the schematic block diagram depicted in FIG. 3. The processing unit 18 analyzes the heart rate signal 22 and the motion signal 24. According to the results of this analysis, the processing unit 18 calculates the heart rate 26 and the speed of movement 28, and the speed of movement 28 indicates the frequency of movement with which part 12 of the body moves during the physical activity of a person 20.

The processing unit 18 further measures the quality 30 of the heart rate signal 22 and the quality 32 of the motion / acceleration signal 24. The quality of the signal 30, 32 shows a measure of the reliability of the data for the measured signals 22, 24. It shows the amount of noise that damages the measured signals 22, 24. Low noise damage leads to high quality of the signal 30, 32, while high noise damage leads to low signal quality 30, 32. The amount of noise damage in the measured signals 22, 24 is measured in the processing unit 18 by performing a frequency analysis, as was explained in detail in the section "Summary of the invention".

Depending on the quality 30 of the heart rate signal 26, the processing unit 18 switches between the two calculation modes. For this purpose, a certain threshold is determined that indicates the noise level in the heart rate signal 22, the heart rate signal 22, which is measured by the heart rate measuring unit 14, becomes unreliable when said predetermined noise level is exceeded, and, accordingly, when the signal reaches the aforementioned threshold. Thus, this threshold means the minimum quality 30 of the heart rate signal 22, which is necessary in order to reliably calculate the heart rate 26 based on the heart rate signal 22.

The processing unit 18 switches to the first calculation mode if the quality 30 of the heart rate signal 22 is above the predetermined threshold. In this case, the processing unit 18 calculates a heart rate 26 based on the heart rate signal 22. This mode is a “normal” mode in which the optical sensor of the heart rate measuring unit 14 provides a reliable signal that includes only a small number of motion artifacts, which still allows calculating the heart rate 26 based on the measured heart rate signal 22 .

If, however, the portable device 10 is subjected to severe shaking (high acceleration), the quality 30 of the heart rate signal 22 may become so poor that reliable calculation of the heart rate 26 based on the heart rate signal 22 becomes technically impossible. Such situations occur when the person 20 moves the body part 12 in a quick and intermittent manner.

In such cases, prior art devices using similar optical heart rate sensors are not able to measure the heart rate, which means that reliable measurement is not possible. However, this problem has been solved by the present invention.

In the cases described above, the processing unit 18 is configured to switch to a second calculation mode in which the heart rate 26 is estimated based on the motion signal 24. This estimate of the heart rate 26 begins as soon as the processing unit 18 recognizes that the quality 30 of the heart rate signal 22 has fallen below a predetermined quality threshold. The processing unit 18 then estimates the heart rate 26 using a physiological model that uses data obtained from the motion signal 24. This assessment in accordance with one embodiment is as follows. At the first stage, the processing unit 18 evaluates the value of HR _constant . This HR _constant value indicates the level of the person’s heart rate to which the person’s heart rate would increase or decrease if the amount and intensity of movement would remain constant over time.

As can be seen from the example diagram shown in FIG. 12, there is a linear relationship between the heart rate 26 and the power of the exercise effort. In this diagram, the X axis shows the runner’s speed, measured in km / h, the left Y axis shows the excess heart rate in beats per minute, and the right Y axis shows exercise power (PAI) expressed in J / min / kg. This graph was obtained from an experiment with approximately 30 participants whose heart rate was measured while running at different speeds and different levels of intensity. It shows that there is a linear relationship between heart rate and the power / intensity with which running is performed. This chart was published in 2005 in European Journal of Clinical Nutrition (2005) 59, 561-570, “Reliability and Validity of the Combined Heart Rate and Movement Sensor Actiheart”.

The experiments performed by the applicant showed that HR _constant more or less equals approximately twice the value of the frequency of movement. Thus, an estimate of HR _constant requires an assessment of the existing movement and, accordingly, an estimate of the frequency of movement of the measured part 12 of the body.

This frequency is obtained by spectral analysis of the motion signal 24. Once the HR _constant value is determined, the processing unit 18 calculates an approximating curve, which is the estimated change in heart rate over time for the time period in which the processing unit 18 is switched to the second mode (for the transition period), in which the optical heart rate sensor 14 the contractions do not give a reliable signal 22. This approximating curve takes the last reliably measured value of the heart rate 26 as its initial value and about enochnoe value HR _constant as its final value. In the interval between these two values, an exponential approximating curve is used. It has been shown that exponential change reflects the natural behavior of the human heart in a good and accurate way.

Since the so-called transition period, in which the optical heart rate sensor 14 cannot measure, is in practice very short, not more than a few seconds, this assumption gives a good approximation. The experiments performed by the applicant additionally showed that the described exponential change in the adaptation of the pulse also depends on the physical fitness of the person 20.

Physical fitness factor a, which indicates the level of physical fitness of person 20, is thus preferably included in the treatment. The experiments showed that an exponential approximating curve, for which the time constant is determined by physical fitness factor a, gives a rather good estimate of the change in heart rate during the transition period. The physical fitness factor a in this case denotes a constant related to the physical fitness of a person in such a way that a decreases with increasing physical fitness 20.

An estimate of the change in heart rate during the transition period can be further improved by taking into account previous measurement sessions of the same user 20. The HR _constant value can thus be determined based on the stored data of previous measurements. In order to realize this, the portable device 10 in accordance with the second embodiment, which is shown in FIG. 4, includes an additional storage unit 34 that is configured to store control measurements of heart rate 26 related to control levels of physical activity intensity person. In accordance with this embodiment, the processing unit 18 is configured to determine the level of intensity of a person’s physical activity based on the generated motion signal 24, as well as determining a heart rate _constant (HR _constant ) by comparing the determined level of intensity with the control measurements stored in said storage unit 34.

Said control measurements may be heart rate measurements that were recorded in previous measurement sessions of the same user 20. If user 20 performed physical activity, such as, for example, running, on the eve of using the same portable device 10, then the measured signals may be stored in the storage unit 34. By recording the heart rate signal 22 together with the corresponding motion signal 24, the calculated heart rate 26 can be connected with corresponding motion levels that are obtained from the motion signal 24. Thus, the processing unit 18, using the previous measurements, is able to determine various constants of the heart rate, which correspond to different levels of intensity of a person’s physical activity.

The intensity level of a person’s physical activity can be determined, for example, based on the frequency of at least one peak value of the generated motion signal 24 and / or on the basis of the average level or amplitude of the generated motion signal 24 over a time interval. The intensity level thus indicates the level of movement, which is a measure of a person’s physical activity.

In practice, the storage unit 34 is implemented by a small microchip. By including an additional memory unit 34, an estimate of the HR _constant value and thus also an estimate of the change in heart rate during the transition period become more effective. The processing unit 18 analyzes the previous recorded data and uses the parameters obtained from these data to evaluate changes in heart rate during the transition period.

As approximately shown in the block diagram depicted in FIG. 4, the portable device 10 may further include a first input interface 36, such as a small keyboard or touchpad, which allows the user 20 to manually select the type of physical activity that he wants to perform. User 20 may, for example, select physical activity from a predefined list, such as running, cycling, rowing, or weightlifting. If the processing unit 18 additionally receives information about the type of physical activity, this simplifies the determination of the frequency of movement. This is again based on the fact that each type of physical activity in practice generates its own characteristic type of structure of the detected motion signal 24.

As further shown in the block diagram depicted in FIG. 4, the portable device may further include a second input interface 38. This second input interface 38 may also be a keyboard or touchpad that user 20 uses to manually enter his / her personal data that physiologically affects the heart rate and its change over time. Possible important personal data may be, for example, resting, age, gender, weight, height and / or resting heart rate.

Thus, the user 20 can directly enter his personal data, which can be used to improve the estimate of the heart rate 26 in the second mode of the processing unit 18, when the optical heart rate sensor 14 cannot provide a reliable heart rate signal 22. The entered personal data can also be used to apply an additional physiological model that can be used to adapt the estimated change in heart rate during the transition period. Exemplary physiological models are known from the above scientific article.

It should be noted that this second input interface 38 is not necessary, since at least some of the personal parameters, such as HR _rest and personal fitness parameter a, can also be obtained from the motion signal 24 in the aforementioned manner. In addition, it should be noted that the first input interface 36 and the second input interface 38 can be implemented by the same keyboard (as is schematically illustrated in FIG. 4).

According to the illustrated second embodiment, the portable device 10 further includes a display 40. The display 40 may be, for example, a small matrix of light emitting diodes (LEDs), which is integrated into the portable device 10 and used to visualize / display the calculated frequency 26 heart rate to user 20 in real time.

FIGS. 5-7 show three exemplary measurements that were recorded using an optical heart rate sensor during user 20 activity. The upper charts (FIGS. 5A, FIG. 6A and FIG. 7A) show a signal 22, 22 ', 22' heart rate measured in beats per minute (Y axis), depending on the time measured in seconds (X axis). The lower diagrams (Fig. 5B, Fig. 6B and Fig. 7B) show the corresponding recorded heart rate quality signal 30, 30 ', 30ʺ and the speed signal quality signal 32, 32', 32ʺ in the respective time periods.

The measurements shown in FIGS. 5A, 6A, and 7A show an uncorrected heart rate signal 22, 22 ′, 22,, which means that these heart rate signals 22, 22 ′, 22ʺ were only measured using the unit 14 optical measurements of heart rate, and were not adjusted by the obtained motion data.

In Fig. 5, it can be noted that the measurement of heart rate was interrupted between 500 s and 700 s (the time period circled). In this time period, the heart rate quality signal 30 is very low and practically tends to zero. Instead, the signal 32 of the quality of the movement speed shows rather high values, which is an indication that there is a large movement of the portable device 10. This leads to an unreliable change in the signal 22 of the heart rate within this period of time.

In this time period, the measured heart rate signal 22 changes very intermittently and very abruptly, which, of course, does not correspond to the “real” behavior of the heart rate. This is based on the aforementioned effect of the appearance in the signal 22 of the heart rate of artifacts of large movements, if high accelerations occur.

Similar examples are shown in Fig.6 and Fig.7. In this document, the failure of the heart rate measuring unit 14 occurs in a period of time from 400 s to 700 s (FIG. 6) or from 100 s to 700 s (FIG. 7). This is identified by the corresponding heart rate quality signals 30 ', 30ʺ, which become very low in these time periods and indicate that the measured heart rate data cannot be relied on.

Figures 8-10 show the same graphs that also show the estimated heart rate signals 23, 23 ', 23ʺ that were calculated / evaluated using the presented portable device 10. Figure 8 relates to the same failure case, as FIG. 5, FIG. 9 refers to the same failure case as FIG. 6, and FIG. 10 relates to the same failure case as FIG. 7. The graphs also include a graph of the "real" 42 heart rate signals that were measured using the ECG device to obtain a real change in heart rate as a reference.

In Figs. 8-10, it can be seen that the heart rate signals 23, 23 ′, 23ʺ estimated using the portable device 10 are very close to the reference signals 42, 42 ′, 42ʺ. It should be noted that heart rate signals 23, 23 ', 23ʺ were estimated based on the motion signal 24 in the manner explained above. In those time periods in which the heart rate measuring unit 14 provides reliable measurement results (for example, in periods of time from 0 to 400 and from 700 to 1850 in the examples shown in FIG. 6 and FIG. 9), signals 23, 23 ', 23ʺ heart rate signals are heart rate signals that are directly measured by the heart rate measuring unit 14. In those time periods in which the heart rate measuring unit 14 cannot provide reliable measurement results (for example, in the period from 400 to 700 s in FIG. 6 and FIG. 9), the heart rate signal 23, 23 ′, 23ʺ estimated based on the motion signal 24 using one of the aforementioned estimation methods. This leads to a fairly realistic calculation / estimation of heart rate at any time during the measurement.

The presented portable device and the corresponding method make it possible to detect the heart rate of an athlete or athlete in many different situations, even in situations in which large movements occur, and the optical heart rate sensor of the prior art is not able to reliably detect the heart rate. As explained above, the processing unit is configured to switch between two modes. In the first mode, the heart rate is calculated based on the heart rate signal if the signal quality is above a predetermined threshold. On the other hand, if the processing unit is not able to reliably calculate the heart rate based on the heart rate signal, the processing unit switches to the second mode in which the heart rate is estimated based on movement by using a model in which data is compared with a set of values, determined from previous measurements or entered by the user. The presented method, thus, uses one of the following features or any combination of them:

1) a history of good measurements right down to the point of failure (unreliable measurement);

2) an indication of the quality of the measurements taken, which indicates that the last measurement is unreliable;

3) a user profile, which can be either pre-recorded or determined from reliable measurements; and / or

4) a physiological model that uses parameters determined from previous measurements for the same user.

As can be seen from FIG. 11, the presented method need not be implemented in a portable device 10. Similarly, a system 100 can be provided that includes a portable heart rate measuring device 44 and a portable motion measuring device 46 that can be turned on. in the same building. The difference between the illustrated system 100 and the portable device 10 is that no portable processing unit 18 is integrated in the portable device. Instead, signals measured by the portable heart rate measuring device 44 and the portable motion measuring device 46 can be transmitted to an external processing device 48, which performs the above calculations / estimates externally. By using a wireless connection between portable devices 44, 46 and processing device 48, this data transmission can also be realized in real time. However, it is also possible that data recorded by portable devices 44, 46 is stored in a storage unit and subsequently transmitted to processing device 48 (after measurement).

In order to establish a real-time connection, portable devices 44, 46 preferably include a communication interface (not shown for simplicity), such as, for example, a radio transmitter, while the processing device also includes a similar communication interface, such as, for example, radio. Otherwise, it should be understood that the system 100 has preferred embodiments similar and / or identical to the claimed portable device 10.

While the present invention has been illustrated and described in detail in the drawings and the preceding description, such illustration and description should be considered illustrative or exemplary, and not limiting; the present invention is not limited to the disclosed embodiments. Other changes to the disclosed embodiments may be understood and made by those skilled in the art in the practice of the claimed invention from the study of the drawings, description and appended claims.

In the claims, the phrase “comprising” does not exclude other elements or steps, and the singular does not exclude a plurality. A single element or other block may fulfill the functions of several elements mentioned in the claims. The fact that certain measures are referred to in mutually different dependent claims does not mean that a combination of these measures cannot be used to obtain benefits.

The computer program may be stored / distributed on a suitable medium, such as an optical storage medium or solid-state medium, supplied jointly or as part of other hardware, but may also be distributed in other forms, for example via the Internet or other wired or wireless telecommunication systems.

Any reference position in the claims should not be construed as limiting the scope.

Claims (27)

1. A portable device for determining the frequency (26) of a person’s heart rate (20), said portable device (10) comprising: a heart rate measuring unit (14) for measuring a person’s heart rate (26) (20) over time in order to generate a heart rate signal (22), - a motion measuring unit (16) for measuring the movements of a part (12) of a human body (20) over time in order to generate a motion signal (24), and - a processing unit (18) that is configured to measure the quality (30) of the heart rate signal (22), calculate the heart rate (26) based on the heart rate signal (22), if said signal quality (30) is higher than a predetermined threshold, and estimates of heart rate (26) based on the motion signal (24) if said signal quality (30) is below said threshold, wherein the processing unit (18) is configured to estimate a heart rate (26) based on a motion signal (24) by estimating a constant heart rate (HR _constant ) and determining an exponential change in heart rate (26) in time, wherein an exponential change in frequency (26) heart rate starts from the last reliably measured heart rate and ends at the estimated HR _constant frequency, which depends on the frequency of the motion signal (24), and the last reliably measured the heart rate is the last heart rate measured by the heart rate measuring unit (14) at the point in time before reaching the threshold. 2. A portable device according to claim 1, in which the heart rate measuring unit (14) comprises an optical sensor, in particular a photoplethysmographic (PPG) sensor for measuring a pulse wave of human blood (20) over time, in order to generate a signal ( 22) heart rate. 3. A portable device according to claim 1, further comprising a display (40) for displaying the calculated heart rate (26). 4. The portable device according to claim 1, wherein the processing unit (18) is configured to determine the quality (30) of the heart rate signal (22) in the frequency domain by analyzing the spectral peaks of the heart rate signal (22) at the heart rate and / or at the frequencies of its harmonics. 5. The portable device according to claim 1, wherein the exponential change is an exponential curve denoting a heart rate over time, the exponential curve having a time constant a , which is a constant related to a person’s physical fitness, which decreases with increasing level of physical fitness . 6. A portable device according to claim 1, in which the processing unit (18) is configured to calculate the value of HR _constant according to the formula HR _constant = 2.1⋅f- a , where f is the frequency of the signal (24) of movement, and a is a constant representing the level of a person’s physical fitness according to the formula a = 75-HR _rest , where HR _rest is the person’s heart rate (20) at rest. 7. A portable device according to claim 1, further comprising a storage unit (34), which is configured to store control measurements of heart rate for known levels of intensity of a person’s physical activity, the processing unit (18) being configured to determine a level of intensity of a person’s physical activity based on the generated signal (24) of movement and with the possibility of determining a constant heart rate (HR _constant ) by comparing a certain level of intensity with control measurements stored in said storage unit (34), if the quality (30) of the heart rate signal (22) is lower than said threshold. 8. A portable device according to claim 7, in which the processing unit (18) is configured to determine the level of intensity of a person’s physical activity based on the frequency of at least one peak value of the generated motion signal (24) and / or based on the average amplitude of the generated signal ( 24) movement in the time interval. 9. A portable device according to claim 1, further comprising a first input interface (36) for obtaining information about the type of physical activity of a person (20), the processing unit (18) being configured to estimate the heart rate (26) based on the signal (24 ) movement and type of physical activity, if the quality (30) of the signal (22) of the heart rate is below the threshold. 10. A portable device according to claim 1, further comprising a second input interface (38) in order to obtain personal data of a person, in particular age, gender, weight, height and / or heart rate at rest, and the processing unit (18 ) is configured to estimate a heart rate (26) based on a motion signal (24) and received personal data if the quality (30) of the heart rate signal (22) is below the threshold. 11. A portable device according to claim 1, further comprising a frequency filter for filtering out those frequency components in the heart rate signal (22) that are generated due to the movement of the device (10), the processing unit (18) being configured to determine the quality ( 30) filtered heart rate signal. 12. A method for determining the frequency (26) of a person’s heart rate (20), comprising the following steps: - measuring a person’s heart rate (26) (20) over time in order to generate a heart rate signal (22), - measuring the movement of part (12) of the human body (20) in time in order to generate a signal (24) of movement, - measuring the quality (30) of the signal (22) heart rate, and - calculating a heart rate (26) based on a heart rate signal (22) if said signal quality (30) is above a predetermined threshold, and an estimate of the heart rate (26) based on the motion signal (24), if said signal quality (30) is below said threshold, wherein the step of estimating the heart rate (26) based on the motion signal (24) is performed by estimating a constant heart rate (HR _constant ) and determining an exponential change in heart rate (26) in time, wherein the exponential change in heart rate (26) begins with the last valid measured heart rate and ends on the estimated frequency HR _constant, which represents the estimated heart rate person (20), which depends on the frequency B Nala (24) movement, and the last valid measured heart rate represents the last heart rate measurement unit (14) measuring the heart rate at the time before reaching said threshold. 13. A system for determining a person’s heart rate (26) (20), said system (100) comprising: a portable heart rate measuring device (44) for measuring a person’s heart rate (26) (20) over time in order to generate a heart rate signal (22), - a portable device (46) measuring motion to measure the movements of part (12) of the human body (20) in time in order to generate a signal (24) of movement, and - a processing device (48) that contains a communication interface for receiving said heart rate signal (22) and said motion signal (24), and processing means that is configured to measure the quality (30) of the heart rate signal (22), calculating a heart rate (26) based on a heart rate signal (22) if said signal quality (30) is above a predetermined threshold, and estimating a heart rate (26) based on a motion signal (24) if said signal quality (30) below said threshold, wherein the processing means is configured to estimate a heart rate (26) based on a motion signal (24) by estimating a constant heart rate (HR _constant ) and determining an exponential change in heart rate (26) in time, with an exponential change in frequency (26) heart rate begins with the last reliably measured heart rate and ends at the estimated HR _constant , which is the estimated human heart rate (20), which depends on the frequency of the motion signal (24), and the last reliably measured heart rate is the last heart rate measured by the heart rate measuring unit (14) at the point in time before reaching the threshold.

Images